Update staging from dev

.env.example

@@ -10,6 +10,10 @@ COMPUTE_ID=compute-env-id
 WORK_SPACE=workspace-id
 WORK_DIR=s3://bucket/workdir
 
+# NCI Gadi project code used for PBS submissions (clusterOptions) across all
+# Nextflow workflows. Optional; defaults to "yz52" if unset.
+GADI_PROJECT=yz52
+
 # AWS S3 Configuration for file uploads
 AWS_ACCESS_KEY_ID=your-access-key-id
 AWS_SECRET_ACCESS_KEY=your-secret-access-key

.github/workflows/check-migrations.yml

@@ -2,7 +2,7 @@ name: Check Alembic Migrations
 
 on:
   pull_request:
-    branches: [main]
+    branches: [dev]
     paths:
       - "alembic/**"
       - "alembic.ini"

.github/workflows/lint.yml

@@ -2,9 +2,9 @@ name: Lint
 
 on:
   push:
-    branches: [main]
+    branches: [dev]
   pull_request:
-    branches: [main]
+    branches: [dev]
 
 jobs:
   ruff:

.github/workflows/test-coverage.yml

@@ -2,9 +2,9 @@ name: Coverage
 
 on:
   push:
-    branches: ["main"]
+    branches: ["dev"]
   pull_request:
-    branches: ["main"]
+    branches: ["dev"]
 
 jobs:
   tests:

alembic/versions/20260717_103718_add_tool_and_require_core_fields_on__7a377a05de6e.py

@@ -0,0 +1,41 @@
+"""add_tool_and_require_core_fields_on_workflows"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = '7a377a05de6e'
+down_revision = '74ab4858580d'
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.add_column('workflows', sa.Column('tool', sa.Text(), nullable=True))
+    op.alter_column('workflows', 'repo_url',
+               existing_type=sa.TEXT(),
+               nullable=False)
+    op.alter_column('workflows', 'default_revision',
+               existing_type=sa.TEXT(),
+               nullable=False)
+    op.alter_column('workflows', 'config_path',
+               existing_type=sa.TEXT(),
+               nullable=False)
+    # ### end Alembic commands ###
+
+
+def downgrade() -> None:
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.alter_column('workflows', 'config_path',
+               existing_type=sa.TEXT(),
+               nullable=True)
+    op.alter_column('workflows', 'default_revision',
+               existing_type=sa.TEXT(),
+               nullable=True)
+    op.alter_column('workflows', 'repo_url',
+               existing_type=sa.TEXT(),
+               nullable=True)
+    op.drop_column('workflows', 'tool')
+    # ### end Alembic commands ###

alembic/versions/20260717_105701_add_unique_constraints_for_workflow__14cff803a3b9.py

@@ -0,0 +1,25 @@
+"""add_unique_constraints_for_workflow_name_tool"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = '14cff803a3b9'
+down_revision = '7616ae331fe5'
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.create_unique_constraint(op.f('uq_workflows_name'), 'workflows', ['name', 'tool'])
+    op.create_index('uq_workflows_name_null_tool', 'workflows', ['name'], unique=True, postgresql_where=sa.text('tool IS NULL'))
+    # ### end Alembic commands ###
+
+
+def downgrade() -> None:
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.drop_index('uq_workflows_name_null_tool', table_name='workflows', postgresql_where=sa.text('tool IS NULL'))
+    op.drop_constraint(op.f('uq_workflows_name'), 'workflows', type_='unique')
+    # ### end Alembic commands ###

alembic/versions/20260717_153902_drop_stale_name_only_unique_index_on__7616ae331fe5.py

@@ -0,0 +1,24 @@
+"""drop_stale_name_only_unique_index_on_workflows"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = '7616ae331fe5'
+down_revision = '7a377a05de6e'
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    # Some environments have a plain unique index named 'uq_workflows_name' on
+    # workflows.name alone, predating this feature and untracked by any prior
+    # migration (likely created out-of-band). It collides with the
+    # same-named (name, tool) constraint the next migration creates.
+    op.execute("DROP INDEX IF EXISTS uq_workflows_name")
+
+
+def downgrade() -> None:
+    # The dropped index was untracked drift, not a schema feature; nothing to restore.
+    pass

alembic/versions/20260724_094600_system_status_incidents_a6ee1af2c156.py

@@ -0,0 +1,33 @@
+"""system_status_incidents"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = 'a6ee1af2c156'
+down_revision = '14cff803a3b9'
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.create_table('system_status_incidents',
+    sa.Column('id', sa.Integer(), autoincrement=True, nullable=False),
+    sa.Column('component', sa.Text(), nullable=False),
+    sa.Column('status', sa.Text(), nullable=False),
+    sa.Column('started_at', sa.DateTime(timezone=True), nullable=False),
+    sa.Column('ended_at', sa.DateTime(timezone=True), nullable=True),
+    sa.Column('message', sa.Text(), nullable=True),
+    sa.PrimaryKeyConstraint('id', name=op.f('pk_system_status_incidents'))
+    )
+    op.create_index('ix_system_status_incidents_component_started_at', 'system_status_incidents', ['component', 'started_at'], unique=False)
+    # ### end Alembic commands ###
+
+
+def downgrade() -> None:
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.drop_index('ix_system_status_incidents_component_started_at', table_name='system_status_incidents')
+    op.drop_table('system_status_incidents')
+    # ### end Alembic commands ###

alembic/versions/20260728_145217_service_usage_4f26ab5038ec.py

@@ -0,0 +1,23 @@
+"""service_usage"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = '4f26ab5038ec'
+down_revision = 'a6ee1af2c156'
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.add_column('workflow_runs', sa.Column('service_usage', sa.Float(), nullable=True))
+    # ### end Alembic commands ###
+
+
+def downgrade() -> None:
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.drop_column('workflow_runs', 'service_usage')
+    # ### end Alembic commands ###

app/db/admin.py

@@ -126,6 +126,7 @@ class WorkflowAdmin(ModelView):
     fields = [
         "id",
         "name",
+        "tool",
         "description",
         "repo_url",
         "default_revision",
@@ -135,6 +136,7 @@ class WorkflowAdmin(ModelView):
 
     _NULLABLE_FIELDS = (
         "description",
+        "tool",
         "prerun_script_path",
     )
 
@@ -158,20 +160,22 @@ async def repr(self, obj: Any, request: Request) -> str:
 
 class WorkflowRunAdmin(ModelView):
     fields = [
-        "id",
+        "submission_timestamp",
+        HasOne("workflow", identity="workflow"),
         "workflow_id",
         "tool",
-        "owner_user_id",
-        HasOne("workflow", identity="workflow"),
         HasOne("owner", identity="app-user"),
+        "owner_user_id",
         "seqera_run_id",
         "run_name",
-        "binder_name",
+        "service_usage",
         JSONField("submitted_form_data"),
+        "binder_name",
         "work_dir",
-        "submission_timestamp",
+        "id",
     ]
-    exclude_fields_from_list = "submitted_form_data"
+    exclude_fields_from_list = ["submitted_form_data"]
+    fields_default_sort = [("submission_timestamp", True)]
 
     async def repr(self, obj: Any, request: Request) -> str:
         return f"{obj.run_name}"

app/db/models/core.py

@@ -7,11 +7,14 @@
     JSON,
     BigInteger,
     DateTime,
+    Float,
     ForeignKey,
+    Index,
     Numeric,
     PrimaryKeyConstraint,
     Text,
     UniqueConstraint,
+    text,
 )
 from sqlalchemy.dialects.postgresql import INET, UUID
 from sqlalchemy.orm import Mapped, Session, mapped_column, relationship
@@ -39,14 +42,32 @@ class AppUser(Base):
 
 class Workflow(Base):
     __tablename__ = "workflows"
+    __table_args__ = (
+        # At most one row per (name, tool) for workflows with tool-specific
+        # configs (e.g. de-novo-design: bindcraft vs rfdiffusion).
+        UniqueConstraint("name", "tool"),
+        # At most one generic (tool-independent) row per name. sqlite_where is
+        # needed alongside postgresql_where so the tests' SQLite schema (built
+        # via Base.metadata.create_all) enforces the same partial uniqueness
+        # as the Postgres migration; without it, SQLite silently drops the
+        # WHERE clause and the index becomes a full unique constraint on name.
+        Index(
+            "uq_workflows_name_null_tool",
+            "name",
+            unique=True,
+            postgresql_where=text("tool IS NULL"),
+            sqlite_where=text("tool IS NULL"),
+        ),
+    )
 
     id: Mapped[UUID] = mapped_column(UUID(as_uuid=True), primary_key=True, default=uuid4)
     name: Mapped[str] = mapped_column(Text, nullable=False)
     description: Mapped[str | None] = mapped_column(Text, nullable=True)
-    repo_url: Mapped[str | None] = mapped_column(Text, nullable=True)
-    default_revision: Mapped[str | None] = mapped_column(Text, nullable=True)
-    config_path: Mapped[str | None] = mapped_column(Text, nullable=True)
+    repo_url: Mapped[str] = mapped_column(Text, nullable=False)
+    default_revision: Mapped[str] = mapped_column(Text, nullable=False)
+    config_path: Mapped[str] = mapped_column(Text, nullable=False)
     prerun_script_path: Mapped[str | None] = mapped_column(Text, nullable=True)
+    tool: Mapped[str | None] = mapped_column(Text, nullable=True)
 
     runs: Mapped[list[WorkflowRun]] = relationship(back_populates="workflow")
 
@@ -72,6 +93,7 @@ class WorkflowRun(Base):
         DateTime(timezone=True), nullable=True
     )
     tool: Mapped[str | None] = mapped_column(Text, nullable=True)
+    service_usage: Mapped[float | None] = mapped_column(Float, nullable=True)
 
     owner: Mapped[AppUser] = relationship(back_populates="workflow_runs")
     workflow: Mapped[Workflow | None] = relationship(back_populates="runs")

app/db/models/system_status.py

@@ -1,9 +1,9 @@
-"""Database-backed cache for runtime system health."""
+"""Database-backed cache and history for runtime system health."""
 
 from datetime import UTC, datetime
 from typing import Any
 
-from sqlalchemy import JSON, DateTime, Text
+from sqlalchemy import JSON, DateTime, Index, Integer, Text
 from sqlalchemy.orm import Mapped, mapped_column
 
 from ...schemas.health import SystemStatus
@@ -38,3 +38,24 @@ def get_status(self) -> SystemStatus:
         Return a SystemStatus object from the cache payload.
         """
         return SystemStatus.model_validate(self.payload)
+
+
+class SystemStatusIncident(Base):
+    """
+    One row per downtime period for a monitored component: opened when a probe
+    first reports ``degraded``/``unhealthy``, closed when it next reports
+    ``healthy``. Healthy checks are never recorded, so storage scales with the
+    number of actual outages rather than with polling frequency.
+    """
+
+    __tablename__ = "system_status_incidents"
+    __table_args__ = (
+        Index("ix_system_status_incidents_component_started_at", "component", "started_at"),
+    )
+
+    id: Mapped[int] = mapped_column(Integer, primary_key=True, autoincrement=True)
+    component: Mapped[str] = mapped_column(Text, nullable=False)
+    status: Mapped[str] = mapped_column(Text, nullable=False)
+    started_at: Mapped[datetime] = mapped_column(DateTime(timezone=True), nullable=False)
+    ended_at: Mapped[datetime | None] = mapped_column(DateTime(timezone=True), nullable=True)
+    message: Mapped[str | None] = mapped_column(Text, nullable=True)

app/routes/system_status.py

@@ -12,11 +12,13 @@
 
 from __future__ import annotations
 
+from datetime import UTC, datetime, timedelta
+
 from fastapi import APIRouter, Depends, Query
 from sqlalchemy.orm import Session
 
 from ..db.admin import require_admin_access
-from ..schemas.health import SystemStatusAdminResponse
+from ..schemas.health import SystemStatusAdminResponse, SystemStatusDowntimeResponse
 from ..services import health
 from .dependencies import get_db
 
@@ -25,6 +27,9 @@
     dependencies=[Depends(require_admin_access)],
 )
 
+# Bounds how far back /system-status/history can be queried.
+_MAX_HISTORY_HOURS = 24 * 30
+
 
 @router.get("/system-status", response_model=SystemStatusAdminResponse)
 async def get_admin_system_status(
@@ -37,3 +42,24 @@ async def get_admin_system_status(
     """Return verbose, admin-only runtime health of the submission components."""
     status_obj = await health.get_system_status(db, force_refresh=refresh)
     return SystemStatusAdminResponse.model_validate(health.to_admin_dict(status_obj))
+
+
+@router.get("/system-status/history", response_model=SystemStatusDowntimeResponse)
+async def get_admin_system_status_history(
+    hours: int = Query(
+        default=24,
+        ge=1,
+        le=_MAX_HISTORY_HOURS,
+        description="How many hours of downtime history to return",
+    ),
+    component: str | None = Query(default=None, description="Restrict to a single component name"),
+    db: Session = Depends(get_db),
+) -> SystemStatusDowntimeResponse:
+    """Return per-component downtime incidents, so admins can see how frequently
+    and for how long a component has been degraded/unhealthy over time."""
+    until = datetime.now(UTC)
+    since = until - timedelta(hours=hours)
+    incidents = health.get_incidents(db, since=since, until=until, component=component)
+    return SystemStatusDowntimeResponse.model_validate(
+        health.to_downtime_dict(incidents, since=since, until=until, hours=hours)
+    )